Behavior of scientific production on severe acute respiratory syndrome

Objective: To analyze the dispersion of journals and scientific collaboration on Severe Acute Respiratory Syndrome, available on the Web of Science. Methods: Bibliometric study that analyzed the production of documents available on the Web of Science, from 1989 to 2020. Results: 1623 documents were retrieved, distributed in 538 journals. The average production per year was 55.9. The average production per author was 1.73. The Nursing category contributed 17 (1.04%) documents, with Virology being the largest contribution 324 (19.96%). China led the ranking with 640 publications (39.43%). Conclusion: The data presented demonstrate the high level of interest of the scientific community in the subject. It was possible to identify 82 authors composing an Elite group of Authors, whose productivity was 66.1%. The United States, China and Saudi Arabia are the countries that most formed Collaboration Networks.


INTRODUCTION
A new virus from the Coronavirus family, associated with Severe Acute Respiratory Syndrome Coronavirus 2 -SARS-CoV-2, whose clinical manifestation confers with severe respiratory disease, appeared in Wuhan, China, in late 2019 [1][2] . We have experienced a century in which information is published and disseminated at an unprecedented scale and speed, making the process of incorporation by researchers increasingly easy 3 , therefore, it is essential to have the minimum ability to select the best indexing bases and repositories, prospecting, filter and select the most relevant information for their research, and it is precisely in this context that bibliometric studies present themselves as sources of information for different research areas, and are capable of subsidizing technological and scientific evolution 4 , however, due to their characteristic of mapping scientific production 5 .
Bibliometric research, which includes the application of statistics to bibliography, has three classically recognized laws: Bradford's Law, Lotka's Law and Zipf's Law [6][7] . The bibliometric studies are based on a set of laws and empirical principles from information science, whose objective is to investigate the quantitative aspects of the production, dissemination and use of available and recorded information, thus contributing to the evaluation of the current state of science, as well as the management of research 6-8 . This study is justified by researching the behavior of the scientific production on SARS-CoV-2, being another contribution to the scientific community interested in studying this serious disease, whose pandemic has led to serious public health, economic and social consequences. The question of this study is: what is the behavior of the scientific production on SARS-CoV-2 in the Science web base? To answer this question through the use of indicators and metrics, the objective is to describe and analyze the behavior of the production of scientific articles on SARS-CoV-2, made available on the Web base os Science.

MATERIALS AND METHODS
Bibliometric study, of a quantitative nature, based on Bradford's Laws -journal dispersion and Lotka -Law of the inverse of the square, to study the behavior of scientific production and the use of electronically recorded information in an international database, published from 1989 to 2020. The analysis material was limited to documents indexed in the Science web base, for this, all types of documents were used as inclusion criteria. There were no exclusion criteria.
The searches were carried out on the Science web database in April 2020 using the descriptor [Severe Acute Respiratory Syndrome] and with a "title" filter. Access was via the CAPES Journal Portal, using the credentials of the researcher at Rede Cafe. The choice of base is due to its acceptance in the world scientific community, especially in the area of health, which provides summaries and citations of peer-reviewed scientific literature, in addition to providing a more comprehensive view of the world's research production. To perform the bibliometric analysis step, the graphical interface of the VOSViewers® software and the Microsoft Excel 2017® tool were used. Figure    Bradford's premise [8][9] is that scientific journals should be organized in decreasing order of productivity of documents on a particular subject, that the whole should be divided into a core of journals more particularly dedicated to the subject and several groups or Zones that should contain the same number of documents as the core, then the number of journals in the core and in the successive Zones, will be the proportion 1:n:n 2 :n 3 ...   1  216  1  216  1  40  2  256  1  32  3  288  1  31  4  319  2  29  6  377  2  24  8  425  Center  1  23  9  448  1  21  10  469  1  18  11  487  1  17  12  504  3  16  15  552  2  15  17  582  2  14  19  610  4  13  23  662  1  12  24  674  2  11  26  696  2  10  28  716  Zone 1  2  9  30  734  2  8  32  750  9  7  41  813  9  6  50  867   17   5  67  952  19  4  86  1028  35  3  121  1133  79  2  200  1291  Zone 2  332  1  532 1623 Source: research data  Transition rates and productivity of authors provide interesting data; the number of occasional or transient authors is lower in disciplines where scientific activity is well established, which means that a high transition rate reflects the lack of relevant working groups, although it may also indicate the presence of researchers from other scientific areas. A transience index of 70.81% was obtained, and from highly productive authors 1.16% was very close to the study on obstetrics 13 , whose values are, respectively, 71.5% and 1.33% of the total.

In this
Applying Lotka's law, it is possible to identify an inversely proportional relationship between the number of authors and articles produced. Nevertheless, the constant found in Applying the Bradford's Law to verify the behavior of the distribution/dispersion of journals, in Table 3, it was possible to identify, from the Bradford Multiplier (Bm), a constant, since the variation between the Bm of the Zones remained within the allowed oscillation [5][6][7][8][9] 0.7, demonstrating a distribution very close to the ideal (I:n:n 2 ), as determined by the Law.
The Nucleus comprised 15 periodicals, being these the most devoted to the subject 9-10 .

CONCLUSION
SARS represents a small proportion of the total scientific production in the health area, however, its coverage in the area was quite representative. Since this is a very current subject, when considering the temporal cut-off, it is reasonable to state that the quantity of documents recovered is quite relevant when compared to other bibliometric studies in the health area.
Researchers can benefit from bibliometric analysis techniques, not only in a quantitative approach, but also in a qualitative one, because the numbers help in the approximate reading of reality and, with the inclusion of more in-depth studies, the richness of the analyses becomes more representative. Therefore, bibliometric studies can contribute to the visualization of the connections between information from several areas of knowledge.
Although based on empirical facts, the laws of Bradford and Lotka, succeeded in this study to confirm possible theoretical hypotheses that the Nucleus of journals is formed by the most devoted and, therefore, more productive, nevertheless, revealed that the greater the specificity of the subject studied, the greater the possibility of identifying groups of authors.
As the classical formulation of the Law was used, it was predictable that the adjustment would not be perfect, since n variables influence the empirical distribution of the literature. These variables are not captured by Bradford's classical formula, and represent the greatest challenge to contemporary scholars of its mathematical formulation. The core pointed out from the application of the Law perfectly recognized an important aspect related to the scientific academic behavior on SARS, that is, when there is evidence that certain journals are or will be recognized in the environment, more and more publications are directed to it, for this, it would be important to verify the qualitative aspect of these journals, based on the impact factor indicators, these would be the most reliable.
Because Lotka's Law provides a platform to measure the productivity patterns of authors over a given period, it is known that each area of a discipline can be associated with an exponent, representing its specific rate of productivity of the author. This is not enough to explain why one individual should be more productive in producing dozens of published articles on a subject, another individual produces few works and a third individual produces none.
The discrepancy in the author's productivity can be partially explained by the context of each individual. For example, author productivity is not linked to collaborative authorship, but collaborative authorship can influence an author's productivity during a given period, as was the case with this research when considering all authors and co-authors, perhaps this can explain in this study, the framing of the law of the inverse square and an Elite Group of authors.
As a contribution, the results of the study and its methodological design may serve as a basis for other propositions, thus filling a gap in the list of bibliometric research in health, especially in nursing, such as the commitment to consolidate its body of knowledge, developing research that can simultaneously contribute to the growth and consolidation of this knowledge.
We highlight some limits to the methodological approach in the present research, including the use of a single research, the difference in sensitivity of the descriptor selected for the search of documents that appear in the title. The exclusivity of these findings may be due to the use of only one database. Thus, future searches may include the examination if the extent of the co-authoring productivity pattern increases if more articles are sampled in different databases. Despite these shortcomings, some results of this study seem interesting and worthy of further discussion.